Venous thromboembolism (VTE) is a major problem in patients with cancer, contributing to mortality and morbidity, interfering with optimum treatment and resulting in high costs. Primary prophylaxis is the most effective management strategy for VTE and is generally recommended for high-risk hospitalized patients. However, cancer treatment and the burden of VTE is shifted increasingly to the ambulatory care setting, but little information is available regarding prophylaxis in cancer outpatients. Successful prophylaxis depends on its application in patients at highest risk, and we have recently developed, validated and published a predictive risk assessment model for VTE in a population of cancer outpatients receiving chemotherapy. In the proposed study, we will build on this model to examine the safety and efficacy of VTE prophylaxis in ambulatory cancer patients and also characterize the contribution of tissue factor to VTE. The proposal is focused on two specific aims. Aim 1 will be to conduct a prospective, randomized clinical trial comparing the safety efficacy of prophylaxis with dalteparin to no treatment in reducing VTE in high-risk ambulatory cancer patients initiating chemotherapy. Patients will be randomized to receive either dalteparin 5000U subcutaneously daily or no prophylaxis for 12 weeks. They will be followed for the development of symptomatic VTE and will undergo screening with serial ultrasonography for asymptomatic DVT and chest CT scan for asymptomatic PE. All thrombotic and bleeding events will be evaluated by blinded adjudicators. Aim 2 will be to establish the value of tissue factor (TF) as a predictive marker for VTE in cancer outpatients receiving chemotherapy. Blood and samples from patients enrolled in the clinical trial described in Aim 1 and from a cohort of concurrent low-risk patients also initiating chemotherapy will be used. We will determine if baseline TF is higher in patients at high-risk compared to those at low or intermediate risk of VTE. Further, we will determine if baseline plasma tissue factor levels are predictive of VTE in the clinical trial. We will determine if increasing TF levels during chemotherapy are predictive of the development of VTE. Finally, we will determine if elevated TF levels correlate with other markers of hemostatic activation. Public Health Relevance: Patients with cancer who are receiving chemotherapy often develop blood clots in the leg veins that cause pain and swelling and can travel to the lungs causing breathing problems or death. We will conduct a clinical trial to determine if preventive treatment with a low dose of a blood thinning drug will reduce the frequency of these clots. We will also do studies to determine if certain abnormalities in the blood clotting system can help predict this problem.